Computational Biology PhD Student
Center for Neural Science
New York University
May 14, 2013
Modeling Fast and Slow Gamma Bands in the Rat Hippocampus
Gamma rhythms in rat hippocampus are thought to be important for a variety of cognitive tasks. When analysing LFP recording from CA1 in the hippocampus, two gamma regimes emerge in distinct parts of the theta cycle: slow (30-50 Hz) and fast (60-90 hz). Here, we propose a model that suggests two types of interneurons and the differing time courses of their post-synaptic effects may be responsible for mediating oscillations in two distinct regimes. Using an integrate and fire model, we show how interplay between the excitatory population and two differing interneuronal populations can cause the system to fall into one or two oscillatory regimes. Our model makes 3 explicit predictions about a network that works in such a way. One, it demonstrates that the slow inhibitory population must significantly limit its firing rate in the case of a fast gamma oscillations, but this need not be true for the fast inhibitory populations in the cases of slow gamma. Two, it shows that the increased input needed to switch a state from slow to fast gamma must act on both the inhibitory and excitatory population, and particularly in must lower the subthreshold voltage to increase the tendency for the fast inhibitory population to fire. 3, it illustrates a necessity for the fast and slow inhibitory populations to interact with each other, and particularly for the fast inhibitory population to have a selective ability to silence the slow inhibitory population. We corroborate this model with physiological network connectivity and interneuronal data from the experimental literature. We suggest that the fast gamma regime is mediated via basket or axoaxonic cells, and slow oscillatory regime mediated via bistratified cells.